一看就懂的Tensorflow实战（Logistic回归模型）

Logistic回归简介

Logistic模型

Logistic模型

Logistic模型图解

损失函数（交叉熵损失）

交叉熵

softmax多分类

softmax

Tensorflow Logistic回归

导入 mnist数据集

代码语言：javascript复制

import tensorflow as tf

# Import MINST data
from tensorflow.examples.tutorials.mnist import input_data
mnist = input_data.read_data_sets("./data/", one_hot=True)

---

代码语言：javascript复制

Extracting ./data/train-images-idx3-ubyte.gz
Extracting ./data/train-labels-idx1-ubyte.gz
Extracting ./data/t10k-images-idx3-ubyte.gz
Extracting ./data/t10k-labels-idx1-ubyte.gz

设置参数

代码语言：javascript复制

# Parameters
learning_rate = 0.01
training_epochs = 25
batch_size = 100
display_step = 1

构建模型

代码语言：javascript复制

# tf Graph Input
x = tf.placeholder(tf.float32, [None, 784]) # mnist data image of shape 28*28=784
y = tf.placeholder(tf.float32, [None, 10]) # 0-9 digits recognition => 10 classes
# Set model weights
W = tf.Variable(tf.zeros([784, 10]))
b = tf.Variable(tf.zeros([10]))
# Construct model
pred = tf.nn.softmax(tf.matmul(x, W)   b) # Softmax

定义损失函数（交叉熵）

代码语言：javascript复制

# Minimize error using cross entropy
cost = tf.reduce_mean(-tf.reduce_sum(y*tf.log(pred), reduction_indices=1))

设置优化器（SGD）

代码语言：javascript复制

# Gradient Descent
optimizer = tf.train.GradientDescentOptimizer(learning_rate).minimize(cost)

训练

代码语言：javascript复制

# Initialize the variables (i.e. assign their default value)
init = tf.global_variables_initializer()
# Start training
with tf.Session() as sess:
    sess.run(init)

    # Training cycle
    for epoch in range(training_epochs):
        avg_cost = 0.
        total_batch = int(mnist.train.num_examples/batch_size)
        # Loop over all batches
        for i in range(total_batch):
            batch_xs, batch_ys = mnist.train.next_batch(batch_size)
            # Fit training using batch data
            _, c = sess.run([optimizer, cost], feed_dict={x: batch_xs,
                                                          y: batch_ys})
            # Compute average loss
            avg_cost  = c / total_batch
        # Display logs per epoch step
        if (epoch 1) % display_step == 0:
            print ("Epoch:", 'd' % (epoch 1), "cost=", "{:.9f}".format(avg_cost))

    print ("Optimization Finished!")

    # Test model
    correct_prediction = tf.equal(tf.argmax(pred, 1), tf.argmax(y, 1))
    # Calculate accuracy for 3000 examples
    accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))
    print ("Accuracy:", accuracy.eval({x: mnist.test.images[:3000], y: mnist.test.labels[:3000]}))

---

代码语言：javascript复制

Epoch: 0001 cost= 1.184004518
Epoch: 0002 cost= 0.665706925
Epoch: 0003 cost= 0.552722147
Epoch: 0004 cost= 0.498915088
Epoch: 0005 cost= 0.465621155
Epoch: 0006 cost= 0.442525349
Epoch: 0007 cost= 0.425387941
Epoch: 0008 cost= 0.412269829
Epoch: 0009 cost= 0.401506593
Epoch: 0010 cost= 0.392485674
Epoch: 0011 cost= 0.384779438
Epoch: 0012 cost= 0.378020378
Epoch: 0013 cost= 0.372379096
Epoch: 0014 cost= 0.367407406
Epoch: 0015 cost= 0.362790742
Epoch: 0016 cost= 0.358502308
Epoch: 0017 cost= 0.354814395
Epoch: 0018 cost= 0.351517630
Epoch: 0019 cost= 0.348201127
Epoch: 0020 cost= 0.345595738
Epoch: 0021 cost= 0.342694492
Epoch: 0022 cost= 0.340168105
Epoch: 0023 cost= 0.338062184
Epoch: 0024 cost= 0.335568684
Epoch: 0025 cost= 0.333836752
Optimization Finished!
Accuracy: 0.8886667

参考

李宏毅机器学习教程

[TensorFlow-Examples]https://github.com/aymericdamien/TensorFlow-Examples

tensorflow

0 人点赞